An educated Martian, looking at the way planet Earth has organized its payment and settlement systems, would find his fingers or sense pods itching to rip them up and start again. Unfortunately these payment systems are not going to benefit from a Martian invasion anytime soon. There is so much vested interest in keeping large chunks of the status quo that change is slow. The main inertia comes from the banks: the beginning and end of practically every transaction, be it online or offline. They are dragging their feet because of the use they make of the free float between order and settlement, and because changing payment systems is expensive.
So bank customers are faced in this electronic age with money-transfer mechanisms that can take three to four days just to make a retail payment from one high street bank to another. Correspondent banking across national borders is a joke. A money transfer can take more than a week and cost $30. Banking practices in a small country such as Finland show that the technology and security are there to make a retail transfer between banks in real time, triggered by a single person-to-person phone call.
Insecure and inflexible
Visa or MasterCard would tell the Martian that they already have an efficient international multi-currency credit-card payment mechanism with hundreds of millions of customers between them. It’s true. But it isn’t secure enough or flexible enough to be the payment mechanism of tomorrow. Today it is costing Visa’s member banks about $50 every time a customer repudiates a payment made over the internet and there is a dispute about it. The repudiations are endemic to barely secure internet payments, in which credit card numbers and details are transmitted over the web. Moreover, the banks have a strange relationship with their creation, the card associations. They are rather embarrassed by the success of MasterCard and Visa and are reluctant to support their gaining any further competitive advantage.
Something has to be done. But it is happening slowly. And the banks are not the prime movers. Out there the Martian sees a Darwinian jungle in which fledgling species of payment systems are moving into the creative vacuum, competing in small pockets of the market, and fighting to gain critical mass.
There are two basic types of e-payment schemes: those that create electronic money (e-money), and those that provide access to a payment mechanism.
The classic e-money scheme is Mondex, which was launched in 1993 and is still trying to establish itself outside the narrow world of shopping centres and university campuses. Mondex uses a stored-value smart card, and is the closest imitation of traditional cash that has yet been devised electronically. Cash can be loaded and unloaded anonymously, person-to-person or from consumer to merchant via the Mondex card, although cardholders can print out details of their last 10 transactions.
Mondex was conceived by a consortium of NatWest, Midland Bank and British Telecom, although in 1997 MasterCard took over 51%. Although the Mondex project is still live and work is being done on TV and PC-based applications that can read Mondex cards, or allow the card to be virtual, it is not being backed as one of the likely winners in the Darwinian struggle. Tim Jones, who pioneered the concept at NatWest, and whose work was eclipsed by the Royal Bank of Scotland takeover last year, still believes it is the best model for electronic cash. But Jones has left NatWest and is spearheading a higher-value payment project dubbed Purseus.
Other electronic money schemes have sprung up in the US, Europe and East Asia. Some have created their own currency, such as Beenz, Flooz, E-Gold, BigVine. They are acting as their own virtual central bank (of which more later). But they are struggling for customers. The limitation of these, and of loyalty schemes that award points to customers who purchase goods, or sometimes who simply click onto a website, is that they rely on building relationships with willing merchants. The websites of such schemes look like rather sad shopping catalogues. Another source of uncertainty is the rate of exchange. Many of these currencies are pegged to the US dollar, but once you’ve bought in there’s no guarantee you can get out. The exception is E-gold, which backs its currency with physical gold, silver or platinum deposited with the Bank of Nova Scotia in Toronto.
Electronic payment schemes are based on national currencies. They are simply data storage and transmission systems that provide an alternative message bridge between bank accounts. Often they rely on loading an electronic wallet from one bank account and unloading the wallet either onto another electronic wallet or into another bank account. In theory, this provides additional security for payments on the web or for credit card payments where the cardholder is not present – such as online or by telephone.
Still seeking the killer app
There is no obvious “killer application” already in operation, although there are plenty of candidates that claim to be the all-singing all-dancing answer to the online merchant’s dream. The truth is that the big banks’ hearts are not in this game, and the banks still control each end of almost every payment. However sophisticated the messaging system in the middle, the account has to be settled by debiting one bank account and crediting another. Nobody wants to keep his money in an electronic wallet where it can do no work.
While banks control the transfer of funds from one account to another they have only a mild incentive to speed up transfers. They have invested millions in interbank clearing and settlement systems. While in Finland, Sweden and Japan there are moves towards real-time retail settlements, most bigger countries have well-established netting and batching procedures for retail payments that they are unlikely to revolutionize in the short term. And there is little incentive to do so. The systems work and they are safe. Although there are some safety arguments in favour of real-time settlement, real-time settlement means banks have to carry more liquidity. For small retail payments the cost of that liquidity cushion may well be prohibitive.
In fact it may turn out that there is little money ever to be made from more efficient retail payment systems, especially from micro-payments (under around $5). The systems may have to be subsidized by advertisers, by users of behavioural data, by the sale of other financial services, or by governments. There is an argument that a safe, small-scale payment system should be a utility, like water, power or sewerage, with an element of public subsidy.
But governments have been reluctant to get involved. A US Federal Reserve committee, reporting in January 1998, explored the possibility of direct involvement by the Fed in the development of an electronic-based retail payment system. “Access to a secure and reliable ‘core’ system for clearing and settling retail transactions – whether provided by the Federal Reserve or by the commercial sector – could be a key element of the growth of emerging payment methods.”
But a successor committee formed in June 1999, the Payments System Development Committee, tended to favour innovation, led by competing private payment systems. Committee chairman Roger Ferguson told a workshop in October 2000: “Successes and failures are bound to occur as the ultimate users of payments systems choose among competing options for making payments. The lesson for public policy is that it should not be built on a single product, system, or vision of the future, no matter how compelling at the time.”
Most of these competing systems have taken care to be “bank agnostic”, that is, not wed themselves too closely to any single bank. Interestingly there is a debate in the UK about creating a payments utility for the socially excluded via the Post office, a so-called “universal bank” – not to be confused with the likes of Deutsche Bank or UBS. The big UK banks are resisting government pressure to be involved – for them it would cost too much and attract the wrong class of customer. However, it seems there would be an opportunity for an online payment system to hook up sub post offices and government benefit departments via the PCs whose installation in the UK’s 18,000 post offices has almost been completed. Those citizens without bank accounts could maintain virtual benefit accounts on the web and perhaps use them to pay their utility bills. This would seem to require government subsidy – the question being how much. The UK government is committed to making all benefit payments electronic by 2003.
Governments were more interested in electronic payment schemes a few years ago when it seemed they might quickly replace cash. That hasn’t happened. The consequence for regulators is that they are more relaxed about the systemic, regulatory and macro-economic implications of e-money.
The attitude of the US authorities has been hands-off – let 1,000 flowers bloom. Fed payment committee chairman Ferguson simply warned in 1998 that good risk management was essential: “The failure of private-sector innovators to address risk early may ultimately force public policy to prescribe solutions.”
New payments systems, new rules
In Europe, the European Commission (EC) decided in the late 1990s to draft a directive to cover electronic money institutions (the EMI directive), which was finally published in October 2000. EU countries are expected to implement it, by harmonizing their own laws or rulebooks with it, by early 2002. The EMI directive appears to have been born out of those earlier concerns that EMIs would grow rapidly and threaten to blunt the traditional instruments of central banking.
The EC’s mission was to set guidelines but also to encourage the growth of small competing solutions outside the banks. Individual EU countries, however, have a spectrum of different views on this, some insisting that EMIs in their jurisdiction should be banks, others intending to use a light touch to exclude as many schemes as possible, such as reward schemes, from even EMI supervision.
The EMI directive says that issuers of electronic money must be supervised as credit institutions, but not as deposit-takers. This is slightly misleading. Most EMIs do take deposits, but they are not deposit-takers in the banking sense: they don’t onlend the deposits as credits at risk. There are strict guidelines on how they can invest the “float” – the money that they take in. Up to 80% of it can be invested in “zero risk” assets such as OECD government bonds; and the remaining 20% must be held in liquid instruments such as cash or sight deposits with top-tier banks.
With such strict guidelines it would seem that there would always be enough assets for immediate liquidation.
But, under the directive, an EMI is also required to have capital and reserves (own funds) of at least 2% of its outstanding issuance of electronic money (either the current amount, or the average of the previous six months).
Some electronic payment schemes, which are likely to have huge intraday volume but minimal overnight balance, are unsure whether they will have to hold capital against their maximum volume or against their average overnight balance. Light regulators may decide such payment systems are not EMIs. At least one EU regulator has privately indicated it will treat as many schemes as possible as falling outside the directive.
But it is clear that electronic payment schemes run a degree of operational risk, against which some capital should be held. Will the charge of 2% be appropriate to cover the risk and/or to encourage EMIs to develop and compete?
Some interested parties complain that the EMI directive won’t create a level playing Weld between EMIs and banks. There are no regulatory restrictions on where banks can invest the float created by an electronic money scheme. If they invest in “zero-weighted” assets such as government bonds, there is zero capital charge for credit risk (although there may be a charge for market value at risk).
A new charge for banks’ operational risk under discussion by the Basel Committee on Banking Supervision may be implemented within the next two years and help to level the playing Weld. But as the rules stand, the minimal capital of e1 million to set up an EMI encourages small non-bank schemes (a bank needs at least e5 million capital). If the EMI’s capital requirement exceeds e5 million (when its balance passes e250 million) it might do better to go for regulation as a bank.
That seems to Wt with the concerns of the European Central Bank, and some of the more conservative European supervisors, that bigger electronic money schemes could make it more difficult for central banks to maintain the stability of their domestic currency, and that they should be supervised as banks.
| Ferguson: good risk management essential | ||||||
Central bankers get nervous
A chapter in the ECB’s November 2000 monthly bulletin worries that multiple electronic money schemes could risk debasing the domestic currency: “Over time, they may be a change in market views about the creditworthiness of electronic money issuers engaging in excessively risky investment activities, which could lead to electronic money instruments being traded at variable exchange rates. This would undermine the role of money in providing a single unit of account as a common financial denominator for the whole economy.”
But European Commission officials and other regulators appear to have little patience with the ECB’s views. (One describes them as “rubbish”.)
ECB officials draw comfort from the requirement in the directive that electronic money, once issued, should be redeemable at par (“in coins and bank notes or by transfer to an account free of charges other than those strictly necessary to carry out that operation”). That seems to suggest that reward or voucher schemes do not fall under the directive. Even a reward scheme such as Beenz, which is backed by the US dollar, is not legally bound to redeem at par, unlike schemes denominated in domestic currency units.
The redeemability factor seems to preclude electronic investment accounts whose portfolios could lose or gain in value. “The EMI would then be engaged in a different activity,” explains an EC official. “But there is nothing to preclude the payment of interest on an electronic money account,” the official adds.
There may be a wide difference in interpretation of the directive from country to country across the European Economic Area. With the “banking passport” system, however, an EMI set up in one jurisdiction, for example Finland or the UK which might be among the more lenient or enlightened, should be able to operate Europe-wide under the supervisory conditions of its home country. Individual countries may set up their own consumer protection barriers, however.
The EMI directive revives the idea of the “narrow bank” – a bank for retail depositors that doesn’t need deposit insurance because its assets are invested at “zero risk” in instruments such as government bonds. The concept grew out of a concern that too many US banks were going bust because of imprudent lending.
The objection to a banking system comprised entirely of narrow banks is that the banks are no longer a source of risk capital. In most economies commercial companies have relied on bank loans funded by retail deposits, transferring short-term deposits into medium-term loans.
So far there seems little likelihood of EMIs accounting for a large share of the deposit market, or of the monetary base. If an EMI were to grow beyond the critical size of e250 million it would probably become a bank and then could make commercial loans – so it would no longer be “narrow”. In the long term, however, EMIs could take over the deposit-taking function performed by banks today, although in the EU at least they are likely to be banks themselves. By that time, say in five to 10 years, the capital requirement for the operational risk run by EMIs, whether standalone or as divisions of banks, is likely to be a uniform 2% of average outstandings.
There is also the view of the EMI as a quasi central bank, since it is an issuer of redeemable digital value. The EMI directive insists that this token value should be redeemable at par, but does not state whether that is in the official domestic currency, multiple currencies, or a commodity or basket of commodities. If the EMI acts as issuer and guarantor of its own currency, it is acting like a central bank. That might be a worry to the European Central Bank, if the currency is tied to the euro yet the issuer is domiciled outside the eurozone.
If you are a central bank, your concerns about electronic money and electronic payment mechanisms may be as follows: That payments may be cleared and settled beyond your control outside the regulated banking or clearing system.
That credit may be issued outside the regulated banking system.
That you have no control over the volume and quality of money issued under a private payment scheme.
That there is a loss of revenue (seigniorage) because of the lower amount of money that needs to be issued by the central bank.
All these factors could affect the powers of a central bank to maintain the stability of interest rates, the exchange rate and the money supply.
At the moment, electronic money schemes are small, and there seems little danger of a private currency, such as Beenz, getting so big that it competes with a domestic currency. Most electronic payment schemes, apart from those like Beenz based on the loyalty-points model, are denominated in major international currencies, and most are settled ultimately by banks.
The float in such electronic money schemes can be regarded as bank equivalent to deposits, and the bank credits, on which the major credit-card schemes are based, are also logged at least monthly as part of the credit aggregates monitored by a central bank.
In the EU the means will be there, when the EMI (electronic money institution) directive is implemented in 2002, to aggregate all EMI schemes into the money-supply figures.
The European Central Bank (ECB) has expressed concern that if an EMI looks shaky and unlikely to redeem its outstandings at par, this could affect confidence in the domestic currency. There does not seem to be much justification for this fear. It would no more dent confidence in the currency than the insolvency of any commercial firm. Unless the EMI was acting as a quasi central bank, issuing and using its standing to back its own private currency; but then only the private currency and those holding it would suffer.
A scheme backed by gold, or a commodity or commodity basket, would be a closed system. If the denomination was a major international currency, there would be the same concerns as there might be, for example, if one of the Scottish banks, which issues its own sterling notes, were to collapse. It would only be a concern for sterling if the Scottish bank’s collapse threatened to become systemic not just an isolated case.
Some worries have been expressed about a scheme that issues value in an international currency, but outside that currency’s domestic jurisdiction. For example a private money scheme in the UK might denominate its payments in euros. Would that be creating euros outside the euro system of central banks? Would it be any different from Argentina, for example, dollarizing its economy? It is a challenge to extrapolate the possible effect of a small payment scheme on an economy, supposing that it was to enjoy exponential growth.
Other concerns have been mentioned – though not by the ECB – that the high velocity of transactions in a successful electronic payment scheme could exert inflationary pressure. The consensus among central bank officials is that you cannot stoke inflation with just an increased velocity of payments: there has to be an absolute increase in the amount of money or credit issued. So far, schemes are too small to have an effect, or they fall inside the regulated banking system.
Even regulated securities and futures exchanges, which see huge intraday activity, are controlled by margin and central clearing house arrangements. One loophole is the over-the-counter derivatives market where initial margin has not always been taken. The over-leveraging of hedge fund Long-Term Capital Management in 1998 exposed that weakness and led to a step-change in the way hedge funds can leverage. So it is unlikely that a non-bank counterparty, such as an unregulated payment scheme, would be able to create enough credit exposure to have a macro-economic effect.
Mobile phones, the new cash
Who will win in the Darwinian battle of electronic payment schemes? That depends on human behaviour, and most predictions along the way have proved wildly optimistic. Human behaviour is changed capriciously by constellations of events and accidents of timing. Take the mobile phone. It is easy to explain now why the social lives of millions of youngsters revolve around this handheld tool. But a prediction of its sweep through Europe and the developing world would have invited ridicule even three years ago. Now some payment experts are confidently predicting the use of SIM cards in mobile phones as the ultimate device for payment ease and security. In years to come shoppers will wave their phone at an item and add it to their shopping basket, walk through a ticket barrier and make a micropayment, sit in a traffic jam and see money unload from their phone as they pollute the atmosphere and go nowhere – a fresh source of frustration and road rage. The mobile phone, waved at a PC using Bluetooth (short-range wireless communication across rooms and through walls), will turn the desktop or laptop into a secure payment terminal, for shopping and securities-dealing on the internet. Biometrics will ensure that the phone user is also its owner.
If that is the Platonic ideal for the future, there are many staging posts on the way. The next stage is the smart card – a card with a chip that allows it to record data to be stored with value, and provide a securely enciphered identity to authenticate the user. Visa, MasterCard, Mondex and Proton (in Belgium) are working on this technology. American Express has already rolled out its blue card, which is a smart card requiring a special card reader to be attached to your PC.
But some payment systems are already ahead of plastic cards and are using virtual wallets in cyberspace. They presuppose that enough people will want to do their shopping via a mobile phone, a palm device, the internet or a digital TV to create a marketplace, or a plethora of marketplaces.
The key to these schemes is critical mass. Either they have to build that themselves, or be sufficiently interoperable to complement and feed on other schemes. Net profit Finance has explored a number of these schemes and assessed the likelihood of their long-term success. It looked at the potential competition for banks from new entrants such as supermarket chains and telecoms companies.
One thing is certain – digital payment will ultimately account for 100% of all business-to-business payments and up to 90% of all retail transactions. The 10% rump will be person-to person-payments between those who like the rustle of paper notes and the near certainty that there is no audit trail.
David Shirreff, former managing editor of Euromoney, now heads Net Profit Finance, a financial research group. It has just published a detailed report : Epayments – the challenge to the financial services industry.
Email: dshirreff@netprofit.co.uk.
